Carnivores – Bobcats

Contributed Oral Presentations


Contributed paper sessions will be available on-demand for the duration of the conference, then again at the conclusion of the conference.


Multi-Scale Occupancy Models of Archery Hunter Observations Demonstrates Forest-Associated Colonization of Bobcats in Illinois
Javan M. Bauder; Kirk W. Stodola; Thomas J. Benson; Craig A. Miller; Maximilian L. Allen
Despite global declines for most carnivores worldwide, some species have expanded their distributions in recent decades. However, documenting broad-scale spatiotemporal patterns in species occurrence is often challenging due to the varying quality of data collected at broad scales. Observations by hunters represent a potentially valuable data source for such efforts, but researchers must account for sampling design, imperfect detection rates, and spatiotemporal variation in detection and occurrence. We used observations by archery deer hunters across Illinois during 2001-2018 in Bayesian multi-scale occupancy models to estimate spatiotemporal expansion of bobcat (Lynx rufus) occupancy. Our objectives were to: 1) estimate county- and site-level occupancy while accounting for imperfect detection, and 2) estimate the extent to which occupancy was associated with county-wide forest cover. Mean county-level occupancy increased markedly from approximately 0.30 to 0.80 during our study whereas mean site-level occupancy showed less of an increase (approximately 0.10 to 0.35). Detection was low (<0.10) but also increased during our study. Both county- and site-level occupancy were positively associated with county-wide forest cover. The positive association with forest cover was reflected in spatiotemporal changes in county-level occupancy. During the early years of the study, bobcats were most likely to occur in counties in southern Illinois and along the Illinois River. By the end of the study, bobcat occupancy was high (>0.80) throughout southern and western Illinois but remained low (<0.40) in the agriculturally dominated east-central parts of the state. Our results highlight the utility of using hunter observation data, analyzed using hierarchical models that account for imperfect detection and multi-scale sampling designs, for monitoring broad-scale spatiotemporal changes in carnivore occupancy. In our case, the analyses tracked the expansion of bobcats across Illinois, but could also be applied to document range contractions of at-risk species.
Spatial Ecology of Bobcats in a High Human Density Landscape of New England
Amy E. Mayer; Thomas J. McGreevy, Jr.; Charles Brown; Brian D. Gerber
Bobcats (Lynx rufus) are a wide-ranging and highly adaptable predator whose populations are increasing throughout much of their natural range. This likely includes the New England states, yet there are only limited empirical ecological studies. How bobcats are responding to the unique modern landscape of southern New England with its highly forested landscape coupled with high human density is unknown. This lack of spatial and population ecological information impedes evaluating recovery and management objectives and identifying necessary management actions. Our objectives were to better understand the spatial structure and distribution of bobcats in Rhode Island. We specifically examined space use, resource selection, and occupancy. We trapped bobcats across 5 field seasons from April 2015 to March 2019, totaling 2,232 trap nights. We captured 8 bobcats and equipped GPS collars to a subset (n= 3) and collected locations for 4 to 9 months. We used GPS locations to estimate annual and seasonal home range size and seasonal resource selection within the home range for each individual. Further, we deployed trail cameras at 100 survey sites during 2 winter and 2 summer seasons. We found the mean winter and summer home range sizes were 219.3 km2and 51.7 km2, respectively. Bobcats selected for forested wetland habitats and were associated with areas closer to wetlands and young forests, according to resource selection models. They also selected for areas with higher road densities, yet avoided developed areas. We detected bobcats at 40 of the 100 trail camera survey locations. Our study highlights the importance of examining a widely distributed species at a local scale in order to employ evidence-based management practices and provides baseline ecological information on an increasing population.
Comparing the Efficacy of Two Immobilization Drug Combinations for the Chemical Restraint of Bobcats
Christopher N. Jacques; Claire C. Helmke; Tyler J. Stratman; Robert W. Klaver
Chemical immobilization of wild animals is used to safely handle wildlife for research, management, and veterinary procedures. We compared the efficacy of two immobilization drug combinations on bobcat (Lynx rufus) bobcats during 2018-2019. We processed 23 bobcats (13 females, 10 males) and alternated between ketamine-xylazine (KX) and pre-compounded BAM; KX consisted of ketamine hydrochloride (HCL; mean = 11.65 mg/kg) and xylazine HCL (mean = 1.78mg/kg) and BAM consisted of butorphanol (mean = 0.91 mg/kg), azaperone (mean = 0.304 mg/kg), and medetomidine (mean = 0.36 mg/kg). We reversed KX and BAM with tolazoline (4.54 mg/kg) and atipamezole (1.56 mg/kg), respectively. We used multivariate analysis of variance (MANOVA) to evaluate potential effects of drug type, sex, weight, and body temperature on post-agonist recovery (agonist to head up [HU], sternal recumbency [SR], standing [ST], and full recovery [FR]) parameters (hereafter, recovery parameters). Our analyses revealed no significant 2-way interactions between drug type, sex, or weight on recovery parameters, thus we reported results only for main effects. Recovery parameters did not differ by sex, weight, or temperature, but was influenced by drug type (Pillai’s λ = 0.99, F2,6 = 28.20, P = 0.03). Time from reversal to ST differed (F1,18 = 10.46, P = 0.005) between KX (mean = 47.40 min, SE = 10.37) and BAM (mean = 10.17 min, SE = 2.07). Similarly, time from agonist to FR varied (F1,19 = 17.38, P < 0.001) between KX (mean = 124.40 min, SE = 12.63) and BAM (mean = 61.83 min, SE = 7.75). Our study provides the first account of BAM for processing bobcats, and provides detailed comparisons of its effectiveness to commonly used KX. BAM was significantly faster than KX and can be reversed instantaneously, thereby providing an efficient alternative to KX for current and future bobcat research and management.
Bobcat Home Range in the Eastern Panhandle of West Virginia
Kirsten Belcher; John Edwards; Rich Rogers; Christopher Rota
Bobcats (Lynx rufus) in North America have historically been harvested as a furbearer species throughout their range. Because they are listed under Appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), State Wildlife Agencies are required to monitor their population status. As part of an effort to improve population estimation of bobcats in West Virginia, our objective was to determine home range size of adult bobcats in the Ridge and Valley Region of West Virginia. During the 2018-2019 trapping season, 21 adult bobcats (F =11, M = 10) were trapped by licensed trappers; fitted with GPS collars; and monitored through breeding (Feb-May), natal care (June-Sept), and winter (Oct-Jan) seasons. Home range was estimated using Minimum Convex Polygon (MCP), Kernel Density Estimate (KDE), and Brownian Bridge Movement Model (BBMM) methods. All methods indicated that the annual home range size of males was significantly larger than females, though estimates varied between methods (MCP, M = 45.36 ±21.95 km2, F = 14.18±12.72 km2; KDE, M = 68.60 km2 ±32.36, F = 21.56 ±16.76 km2; BBMM, M = 45.43±15.23 km2, F = 14.11±9.92 km2). The male home range estimate produced by the KDE method was significantly larger than the male home range estimates produced by the other methods. There was no difference in home range estimates of females between methods. Seasonal differences in home range sizes were not found for either sex, regardless of estimation method. Our findings will improve population models for bobcat in West Virginia and inform management decisions including future trapping regulations.
Estimating Density of Bobcats in West-Central Illinois Landscapes Using Spatial Capture-Recapture
Tyler J. Stratman; Claire C. Helmke; Robert W. Klaver; Christopher N. Jacques
As a prerequisite to successful management and monitoring, there is considerable need to understand bobcat (Lynx rufus) density and abundance throughout fragmented Midwestern landscapes. Spatial capture-recapture (SCR) modeling has become widely applied in ecology as it provides a rigorous framework for estimating density, while accounting for heterogeneous space use related to the spatial organization of individuals relative to camera traps. We aim to improve the use of remotely triggered cameras as a viable non-invasive sampling technique to estimate density of bobcat populations in fragmented landscapes. We deployed paired camera-trap stations (n=225) in Fulton, Hancock, Schuyler, and McDonough counties of west-central Illinois during January-April (i.e., bobcat breeding season) of 2018 and 2019. We used the oSCR package in Program R concurrent with radio telemetry data from previously collared bobcats, to estimate density as a function of space use. We collected over 870,000 images from 225 paired-camera stations. Total number of bobcat detection events during 2018 and 2019 were 238 and 196, respectively. During 2018, we captured 59 uniquely identifiable bobcats 172 times and recaptured these same individuals 113 times; 15 of 59 (25.4%) were recaptured at >2 camera stations. Similarly, we captured 51 uniquely identifiable bobcats 113 times and recaptured these individuals 62 times during 2019; 17 of 51 (33.3%) were recaptured at >2 camera stations. We estimated bobcat density at 6.36 individuals (95% CI = 4.86-8.32)/100km2 and 4.77 individuals (95% CI = 3.58-6.37)/100km2 in 2018 and 2019, respectively. Our analyses revealed a marked increase in density estimates from previous SCR-derived estimates of bobcat density in Illinois, which may reflect structural differences between density estimators or increasing bobcat abundance across Illinois. Nevertheless, our revised density estimators are timely and may aid in future bobcat management and conservation decisions across Midwestern landscapes.
Survival of Bobcats in a Newly Harvested Population in West-Central Illinois
Christopher Jacques; Robert Klaver; Tim Swearingen; Sean E. Jenkins; Stan McTaggert
Increased understanding of survival and causes of mortaility of bobcats (Lynx rufus) is a prerequisite to successful management programs. Bobcat survival has been quantified throughout most of its geographic range, though few studies exist of newly harvested populations in Midwestern landscapes characterized by relatively high human activities. During 2016-2020, we captured and radiocollared 44 bobcats (23 M, 21 F) in westcentral Illinois. Pooled annual survival rates for kitten, juvenile, and adult bobcats was 0.96 0.89 , and 0.67, respectively; pooled annual survival across all years, sexes, and age classes was 0.82. We documented 16 deaths, most (87.5%) of which were human-caused and primarily attributed to vehicular collisions and harvest. Distribution of mortalities differed by sex (χ21 = 10.47, P < 0.001), with more deaths occurring in males (n = 11) than females (n = 5). Despite their vulnerability to anthropogenic mortality, human influence is not currently limiting bobcat populations in westcentral Illinois. If human population growth in rural areas continues to decline, bobcat populations in westcentral Illinois may be less vulnerable to human-caused mortality. Nevertheless, population managers need to account for the cumulative affects of total mortality when implementing harvest limits on newly harvested bobcat populations, which also may aid in facilitating the expansion of bobcat populations into fragmented Midwestern landscapes as compared to other less fragmented landscapes throughout the species range. Wildlife managers in adjacent Midwestern states (e.g, Iowa, Indiana) could use our age-specific or pooled annual estimates to model survival in landscapes with similar road densities, landscape features, or human population densities.
Implications of Road Mortality for the Population Viability of Recovering Bobcats in Ohio
Viorel D. Popescu; Marissa Dyck; Kevin Shoemaker; Maddy Back; Ryan Brown; Madeline Kenyon; Catherine Dennison
Understanding and documenting in wildlife populations trajectories are important for sustainable wildlife management and conservation of wildlife. Bobcats (Lynx rufus) are currently recovering in Ohio from local extinction due to habitat loss and overharvesting. The aim of this study is to determine the viability of the recovering bobcat population in Southeast Ohio using non-invasive techniques and drawing on multiple data sources. Currently, road kill is the main source of mortality, estimated at a minimum of 6% (and up to 15%) of the Ohio population. Here, we draw on a road kill dataset collected by Ohio DNR in 2012-2014 and 2019-2020 (~300 animals) to evaluate the demographics of road kills animals, and infer the effects of road mortality on population trajectory. We used fecundities based on necropsied Ohio road kill animals and literature, and survival rates from an Ohio telemetry study, road mortality, and literature, to build a spatial age-based population model. Fecundities were similar to those in other Midwest populations, and we found that bobcat road kill was skewed towards young age classes, suggesting that additional sources of mortality, such as legal trapping or poaching may be additive. The spatial component of the population model was based on a 2nd order selection habitat suitability model developed from 1000+ verified sightings, which suggested that a potential limiting factor to population expansion is habitat availability (forest and forest/open habitat mosaics). Overall, a positive population growth rate was sustained despite high road kill levels and high dispersal abilities facilitated colonization of most forested areas of the state, but future management decisions needs to consider impacts on vital rates with high sensitivity (adult and kitten survival) and potential habitat limitations.
Influence of Landscape Metrics on Home Range Use by Bobcats in Fragmented Midwestern Landscapes
Claire C. Helmke; Tyler J. Stratman; Robert W. Klaver; Christopher N. Jacques
Increased understanding of space use requirements of bobcats (Lynx rufus) across fragmented landscapes is a prerequisite to successful management programs. A fundamental conservation issue pertaining to this species has been to what extent habitat fragmentation affects variation in home range size and fine-scale habitat use across the midwestern United States. Thus, our objective was to quantify potential effects of habitat on home range use by bobcats in agriculturally dominated landscapes of west-central Illinois. We calculated 95% and 50% adaptive kernel home range and core areas, respectively, of adult bobcats from 2017-2019. In addition, we modeled potential effects of landscape and class-level habitat metrics on bobcat home range use. Our analyses revealed significant differences (F1,32 ≥ 15.17, P < 0.001) in composite 95% home range and 50% core areas between male (x = 109.00 km2 and 23.66 km2, respectively; n=13) and female (x = 33.65 km2 and 8.03 km2, respectively; n=17) bobcats. Composite 95% home range size was best described by a model (AICc weight = 0.79, R2 = 0.39) containing forest patch size (β = -3.10, SE = 1.64, 95% CI = -8.57 to -1.43) and distance between water bodies(β = 0.06, SE = 0.02, 95% CI = 0.02-0.10), thereby confirming the importance of forested habitat in relative close proximity to water as primary habitat features selected by bobcats in fragmented landscapes. Our results have future management implications in relation to expected habitat use, recolonization rates, and continued population expansion of bobcats across Midwestern landscapes compared to other areas throughout the species geographic range where habitat quality is higher and land use is less intensive.
A Longitudinal Study of Shifting Habitat Selection by American Martens in Response to 30 Years of Intensive Forest Harvesting
Tyler F. Woollard; Daniel J. Harrison; Erin M. Simons-Legaard; Kirstin E. Fagan
Understanding how patterns of habitat selection change in response to the cumulative effects of anthropogenic disturbance can provide insight into the adaptive capacity of populations to persist in human-altered landscapes. We evaluated temporal patterns of patch-scale habitat selection by American martens (Martes americana) occupying commercial timberlands in Maine across a 30-year period characterized by habitat changes resulting from forest harvesting. We monitored 115 resident, non-juvenile (>1 year) martens (63 male, 52 female) spanning 143 marten-years during three sampling intervals (1989-1990, 1994-1997, 2018-2019). Within home ranges, median availability of tall (>12m tree height) uncut forest decreased from 62.3% to 20.1% from 1989-2019. Availability of tall (>12m tree height), older (≥ 34 years post-harvest) regenerated clearcuts and partial harvests increased from 0% to 19.5% and 24.6%, respectively. Marten consistently exhibited strong, positive selection for tall uncut forest. In contrast, recent (< 6 m tree height) and regenerating (6m - 9m tree height) clearcuts were consistently avoided, but selection for older regenerated clearcuts was similar to tall uncut forest after trees exceeded 12m. Selection for partial harvests, which represent the majority of contemporary harvests, was consistent and equivocal. Marten use of tall uncut forest exceeded availability across all intervals, but use decreased through time as availability declined. Coincidentally, use and availability of older regenerated clearcuts and partial harvests increased through time. Marten appear to have adapted their patterns of patch-scale habitat use in response to the progressive decline in availability of their preferred habitat, tall uncut forest, by shifting use towards older regenerated clearcuts and partial harvests. Habitat use is not synonymous with individual performance, however, and we suggest that future habitat selection studies consider responses in relative individual fitness of species associated with mature forest to increasing fragmentation and the cumulative effects of forest harvesting.


Location: Virtual Date: Time: -